Regional Buckets are an experimental, innovative, and rapidly changing
new feature for Google Cloud Storage. Unfortunately, being on the bleeding edge
means that we may make backwards-incompatible changes to Regional Buckets,
and the Google Cloud Storage SLA does not apply to Regional Buckets. We'll
inform the community when this feature is no longer experimental, at which
time we'll provide an SLA for Regional Buckets.
Regional Buckets allow you to place your Durable Reduced Availability data within the same region as your Compute Engine instances.
How Regional Buckets Work
The new locations for Regional Buckets allow you to create your bucket in the same region as your Compute Engine instances. All Cloud Storage and Compute Engine instances in a region share the same network fabric, reducing the latency between your Compute Engine instances and your objects and increasing the bandwidth. This gives you better performance for data-intensive computations, as opposed to the existing storage offerings, which spread your data over a much larger region to keep it closer to your users.
Objects placed in these buckets will be kept in the region on a best-effort basis. If necessary, the data may be kept elsewhere in the United States and moved back as capacity allows. You will not be able to directly query the exact data location. Since there are only a limited number of datacenters in each of these regions, only Durable Reduced Availability storage is available.
- The location where your bucket and the objects in it are stored.
Locations may be specified at a national/continental level (US, EU, ASIA) or at a
- The region describes a more specific geographic location for your resources. When using Regional Bucket, choose a region that makes sense for your scenario. Usually, this will be the same region as the Compute Engine zones containing your instances. A region contains one or more highly-connected zones.
- A zone is an isolated location within a region that is independent of other zones in the same region. You use zones to specify where your Compute Engine instances are deployed.
Creating a Regional Bucket
To create a Regional Bucket, specify one of the new regional locations instead of US, EU or ASIA. The regional locations are:
You can pass the location to the
gsutil utility, for
gsutil mb -c DRA -l US-CENTRAL1 gs://myregionalbucket
Note that you must specify "-c DRA" because only Durable Reduced Availability storage is available for regional buckets.
You could also specify that you want to create a regional bucket in an XML or JSON PUT request, in the location and storageClass parameters.
Using a Regional Bucket
To use a Regional Bucket, simply reference it from your application as you would any other bucket. No special API calls are required after you have created the bucket.